On The Brink with Castle Island - Luuk Strijers (Deribit) on a New Model for Proof of Reserves (EP.387)

Episode Date: January 17, 2023

Deribit Chief Commercial Officer Luuk Strijers joins the show to talk about their new Proof of Reserve rollout. In this episode:  Why Deribit decided to do a PoR  What were the considerations going into it? What made Deribit cautious about the procedure? Which PoRs did Deribit look at for inspiration? Has Deribit received client feedback from the procedure so far? How does Deribit demonstrate asset ownership? Unique among exchanges, Deribit was able to do a PoR for all of your assets, and on a daily basis. How did they manage this?  Why Deribit chose to release the full list of liabilities  How does Deribit achieve end user privacy given the full liability dump? What data, if any, is leaked by the attestation? Did you look at audit firms in terms of supervising the PoR? Why did you go the auditless approach? Does Luuk anticipate moving to ZK proofs for liabilities eventually? How can PoR be improved?  Related episodes:  OTB 180, The Auditor view of PoR  OTB 212, The Proof of Reserve Restoration  OTB 283, Jeremy Welch (Kraken) on Contemporary Proofs of Reserve Related content:  Nic Carter on Medium, The Status of Proof of Reserve as of Year End 2022

Hello and welcome back to On the Brink. As we escape the wreckage of 2022 into the bright horizons of 2023, there are a few bright spots to look back on, namely exchanges beginning to compete on credibility and trust as opposed to just product listings. One of the interesting phenomena that's occurred is the emergence of proof reserves. This has been done by exchanges since 2014 on and off, but proof reserves now has more traction than it has ever had historically. In the last couple months, 10 major exchanges have done proof reserves out of stations. Each of them slightly different from the other. One of the best ones is Deribut.

So they followed a model where they're releasing all of the client liabilities, of course with some privacy protections built in. They're able to achieve universal coverage on their asset base. and a frequency of once daily, which is really excellent relative to a lot of their peers. Now, that model isn't suitable for every exchange, but Deribut was able to make it work, and this provides their clients a really high level of assurance that they actually have the assets that they say they have. So today I'm sitting down with Luke's three years, the chief commercial officer at Deribut, to dive into why they chose to do their proof reserve and how they're able to accomplish it and reach such a high standard.

It's time running. Hello and welcome back to On the Brink. I'm Nick Carter. This is yet another proof of reserves episode. One of many proof reserves is taking off. Thanks in large part to the efforts of folks like my guest today, Luke's Thraer, who is the chief commercial officer at Deribut. Luke, thank you so much for joining us.

Thank you for the invites. Glad to be here. So, as you know, I did a little review of all the proofs of reserve that happened at the exchanges and Deribut got my joint top score. So congratulations for doing that. I mean, I was actually very impressed by the proof reserve that you guys put together. So thank you. And it's it's good to be recognized. We tried, uh, we tried all we could to find like the best outcome for us, our clients and to be as transparent as possible. But it's, it's nice to be recognized for those efforts. So it's more work than it seems. So it's, it's,

It's not if someone actually looks and compares and makes an assessment of how good certain approaches are. Yeah, I can't imagine there's too many proof reserve enthusiasts out there, but there's at least one. So, you know. Yes, indeed. So proof reserve, it is kind of a cost center, right? Like it is something that takes a lot of work and has certain tradeoffs in terms of, you know, operational difficulties, exposing keys or at least requiring that you make transactions and exposing some data, although we'll dive into the client data that it's not client data, but we'll dive into

what is exposed or leaked, but it's still kind of worthwhile. So what was the calculus for you guys in terms of deciding to jump in and get started with the proof reserve? I think it's an overall trend. So if you look at, let's say a few years ago, 2019 when institutions started coming in. And those are not the big investment banks, but since that year we've seen like 1,500 companies join,

which is an amazing trend, but over time, initially it was those companies onboarding with us. And over time, the companies got bigger and it's us onboarding with them as well. So instead of just us asking for proof of residence, passports, incorporation documents, those companies ask us for the same. So we have to file all kinds of due diligence requests and information sharing. And I see Prover Reserves as the next step.

People want transparency. They want to trust the platforms where they're putting their assets, especially as we're a centralized platform. So they want to make sure that they know who the people are, that they know the tech set up, that they know the finance, the financial health. that they know, et cetera. So this is simply a next step in transparency. And we've, I think, hold off, held off for a bit,

also because of security reasons, you don't want to share everything. But at some point after June, November, FTX, June, three hours, we realized that this trend has been set in motion. And the next step would be wallet and liability transparency. So we immediately agreed that that's the next step. We will fine because we've been transparent in everything, essentially. So our track record is quite public. We're not perfect, but we have, I think, always been friendly towards clients.

And we've been transparent in our mistakes. So if something went wrong, we've always explained. We've always compensated people impacted. This is the next step. So again, we want to be at the forefront of transparency. And so we immediately decided to publish our wallets, but that's only half of the work. The wallet publication is simple because that's just a list. But the next step, the liabilities, that takes more time.

And then the next logic would be simply matching those. Now, was this something that clients had actually asked you about? Or was it just to kind of your reaction to some of the failures in the exchange space? It's both. So people ask from a judilious perspective, not for your full wallet list, but for your hot wallets and they want to monitor transactions. So we sometimes get the question, can you explain this transaction that happened in October, 2018 and whatever? So over time, of course, stuff happens and people might get hacked, so not the exchange, but people. So if a transaction is done from Derbid to whatever, some wallet of someone who shouldn't be the owner of those assets and those assets are compromised in a way, of course, people doing their research want to know why did we do this and etc.

So from a judicious perspective, we occasionally get questions about our hot wallets and the aim is not checking the balance, but the aim is checking the activity. So that happened over time. And that's rather normal, I would say. But the next step would be checking the assets. And that's what's happening now. So that's the last months or after FTCX, when people realized it's strange that FDX didn't have the assets they were supposed to have. They want to make sure that that won't happen again.

And when you were evaluating the existing proof of reserve landscape, were there specific, because there's a lot of heterogeneity among the proceedings? Were there ones that you took inspiration from that you'd already seen in the market? No, we were already, there were some papers out there about mercilaries and the proof reserves approaches that were recommended. So we were aware of those already. We are not Binance, we are not Coinbase, we don't have a hundred million clients. So in terms of file size and complexity, we always thought a full mercantry is not necessary because in a mercantry, Mercantry essentially you would show the sum of the leaves which would reduce the file size.

And we said, okay, we don't, as we don't have 100 million clients, and especially if you cut up balances in multiple smaller buckets, it would mean for, let's say for Binance with whatever, a thousand assets or 100 assets, if you cut all of those into pieces, we're talking about the hundreds of of millions or billions of entries. If a client has positions in 10 coins and you want to cut to all of these, one client could be 100 entries already. So for Bynes, publishing everything would be file size wise, almost impossible and it would be terrible to handle.

So that's why some of these large platforms only take a subset of assets and take the mercenary approach. For us, it's simpler. We have less assets and we have less clients. So we have very big clients, but small ones. And for that reason, we said, why don't we just aim for full transparency

instead of a derived version? And that's what we did. So it's the, I think the, the cleanest version, but only possible because we have less clients. So that's why we have the ability to do what we did. Yeah, it's certainly easier to reason about with the full liability release. Yeah.

So you've been undertaking the procedure on a data, basis since mid-December is that right we started yeah somewhere around that yeah and so have you actually received client feedback I mean is this something that people are excited about the ones that are excited are very limited it's you with us and I think it puts them at ease so not having it would get them less excited so the fact that we have it the fact that someone like you takes the time to review, make sure that this actually works and some clients check. So the bigger clients do an occasional check and ask how to interpret something.

But the rest, I think, is at ease because we do it on the daily basis. And it is fully transparent. And there's people out there that review stuff and ask some questions and are critical towards platforms that are doing less of a job. and in the absence of that criticism, I think that that pleases them. So it's good to have. It's like an insurance policy.

It's nice to have a fire insurance, even though you never have a fire, but still it's nice to have one. And you don't look at all the details. You assume the fire insurance would pay when there's a fire. It's a bit like this. The fact that we have it, and we've taken an effort to make available the best,

version out there. So we have all the FACIS, we have some full liabilities and we have a daily file. So that there's nothing more that we can do except for signing the wallets, which we have started doing this week and we've signed half of them. So that that showcases full ownership and control of the wallets. Yeah, that's about it. And I think people are pleased that we do it. Some people ask questions. Most people just say thank you and appreciate your efforts. Yeah. I mean, I think Prove Reserves. is quite revolutionary, actually. And me too.

I don't think most people get excited by kind of accounting procedures. But I mean, if you think about this relative to traditional assets, it's not possible to cryptographically prove ownership of some financial, some non-crypto financial asset to a third party. That's just, I mean, you think about gold, custody, other kind of store value assets. you just have to really just take it on faith that your gold custodian has the gold. And a lot of people have questions about whether some of these entities have the gold that they claim they have. So this is one of those things where I do see the crypto space is actually, of course, we've had issues with custody in crypto, historically.

No question about that. This is one of those situations where we're actually now creating a level of trust and assurance, which exceeds that available in the traditional finance space. It's an amazing learning curve and I fully agree. So it might be boring, but it is revolutionary from a trust perspective. So the insights platforms like us are able to give on a wallet level, on the liability level, on a timing level, because we do it daily, but you could even do it like hourly or real time if we wanted to do. The benefits of that from an equality assessment, do you trust this platform?

Do they do weird stuff in the past? Blockchain is already extremely transparent, but this next iteration of the implementation of blockchains is I think it's really special. If you look at traditional finance or you look at banks and you look at custody, you mentioned gold, but you can mention the same for shares. Are those shares really there? You buy some Tesla shares. Does your broker have those shares or are they on borrow? Are they on somewhere else? Whatever. All of that stuff is impossible in this next version. So yes, I like where we're heading. And I think the next step, and that's the question of how to get there, would be financial health. So if you look at the assets are there, you still don't know the health of the company.

So we've appointed, and it's really expensive, we've appointed a top four auditor, and we're in the audit readiness phase, which means it doesn't mean that we are audited yet, but it's in the preparation phase, which is more than a thousand hours of work already for them to assess all the policies, procedures, etc. But the next step would be if we are audited, we have an annual report, which is nice, but it only published, whatever, four months after the year end. So it doesn't mean that much yet. It doesn't mean that your policies are vetted. It means security is vetted. It means a lot of things. So it's definitely a very good stamp of quality approval. but you also want more real-time insights into the health of the company.

And I think that's the missing opponent still. So you want to have assets, you have liabilities. You want tech assessment. So we are publishing later this month a penetration test. So we've appointed an external company. We do this every year. But now we've taken a public version. So typically you can choose whether you want this penetration test results to be public or private.

If it's public, it's a differently written report showcasing the findings. So we will share that. So then you have a technical assessment. You have your assets assessed. You have your liabilities assessed. If we are audited, you have your financial health assessed. We've made available margins locked. I think we're the only platform that does that, which means that you can see how many assets we have.

we show how many open interests we have. And now you can also see how much margin do we take for all of this open interest, which is again, a next level of transparency. Yeah, I mean, that was one of the problems with FTX was not just that they didn't have the assets. I mean, they had a lot of problems, but also the margins didn't judge the margins or, yeah. Yeah, there was no transparency around that and then ultimately was the big margin position, which took them down.

Yes. So, and that should be impossible if you have. plenty of margins, any move, and the market moves that took them down were not that big, actually. So we look at scenarios, right? So we have, what happens if the money, so we real time check every market move up 15% and down 15% and volshocks. So we look at your full portfolio in buckets of 36, 9, 12, 15 and down the same in terms of price moves. And then within these these buckets, we look at volshocks. So what happens if volatility moves up 40% or down 40%

and the price moves up a certain or down a certain way? And we look at all of that real time and we assess the impact for your full portfolio should that scenario happen. And the worst of all of these scenarios, that's the margin we charge. But we also look beyond that. We look at what happens if the market moves 50% down. Of course, that's going to result in carnage. but we tried to quantify this carnage and how many liquidases would there be

and which clients would be resulting in problems, etc. And in the FDX scenario, I think that's all omitted. That wasn't properly done or they exclude some clients in those assessments, which one of them would be themselves. But at least step one would be the transparency for the margins. So you can see we have something like 7 billion open interest at the moment. you can see exactly how much margins we charge for that open interest. And you can also see if the market moves up 10% or down,

what happens to those margins? Are they still there? Are we liquidating or whatever? So it's fully transparent. And I think the last missing component or components would be custody of choice. So we have the proof reserve model is deribate custody, right? So we are custodian. But for us, we offer external custody as well.

So we have copper and we have cobo. And we're adding a third one, hopefully this month or next month, allowing clients to choose. So I believe is firmly that in the end, and we are an institutional platform. So institutions have their preference. So someone from Asia prefers an Asian custodian, someone in the whatever, the asset management space, prefers a custodian that's experienced in whatever reports they want and exactly how they do it for equities they want it for crypto as well. So we believe that in the end, it should not be the platform determining where to custodize. It should be the client's choice. So that's why we've

always been at the forefront of adding those. And that's why we are, I think, the only one that is adding a third one now, which means that pilots can choose. That's complete freedom. So we would lock assets within that custodian. The clients can see that those assets are there or the custodian can tell them, look, those assets are there and the proof of reserves, responsibilities move from the platform to the custodian and each of them have their own solutions, whether it's on chain or off change, so not every custody solution is the same. And we would simply lock within that external party and we would move P&L on the daily basis. So actually, that was going to be my next was I do see the same thing happening, which is the segregation of exchange and custody.

I mean, that's what we have in traditional markets. I think it's happening in crypto, especially as, you know, some certain clients become more leery about collateralizing with the exchange directly. And especially as the quality of custodians improves, I think you'll continue to see that. So in the case of your proof of reserve, how did you take that into account? We take them out. So we take them out on two sides. So if you are, so, One of the bigger ones we service is copper, copper clear loop. So let's say you have 10 BTC in copper. We don't exclude or we don't include your account in the liability side because we can't.

We're not responsible for your 10 BTC. We can't show proof that we control them because we don't. So the liability accounts, the copper ones are excluded. They don't see themselves in the proof reserves and the assets are not included either because it's, we have direct controlled accounts and not the copper accounts. So our assets are actually, I think something like 5,000 BTC higher or 6,000 something around those areas because we have those places at external custodians. But the liabilities are also understated. But there's no alternative because it should be the custodian showing proof of reserves instead of the platform.

Agreed. Yeah. And hopefully we get to a point where ProV Reserve is normalized across the whole space and then the custodians themselves do it and so they can pass that on to the exchanges such as yourselves. If you are an on-chain solution, it's simple, right? We use Firelocks ourselves. It's all on-chain. So we can just show a fireworks address and that's it. If you are an off-chain solution, it's more complex because it might be non-dedded accounts per client. It might be omnibus status where. they simply keep access in bulk instead of segregated. But yeah, so in the end, we said, okay, our direct assets are a certain size, and those are matched by the liabilities or the other way around. We have so many liabilities which is matched on chain by certain assets. So uniquely among the exchanges that I profiled, I think I looked at 10 exchanges that had done a proof of reserve since November of last year,

you guys were able to achieve, I think, 100% asset coverage in terms of your proof reserve. I think it's because you have fewer collateral types in most exchanges. Were there any complexities on a per asset basis? No, no, really, because we don't have to a simpler platform than some of the others, but we don't do marginate and we don't do cross-collateral. So I can imagine that, let's say you have cross-collateral and you deposit a million U.S.D. and you buy in, let's say, an inverse perpetual, BTC perpetual, and you lose 10 BTC. If you lose that 10 BTC, it's not there.

So in the end, you have to sell some use DC and buy BTC to cover the other side of the equation. So if you lose 10 BTC, someone else gains 10 BTC, and this 10 BTC, doesn't exist yet in that setup until you offset the UCC for the BDC. And I think that's, or I hope that's part of the reasons why other platforms are under-collateralized because I can't find another reason. For us, it's simple. You deposit 10BDC and we need to have 10 BDC because otherwise there's a problem. We have 100% of the assets and if there's a bank run, it would be unfortunate, but there would not be a problem.

The only problem we face is access. So not all wallets are equally accessible. So for some of them, we need to go to a vault, to a bank vault. And if the bank happens to be closed, we have a problem. And the problem is not that the assets are not there. The problem is time. So the assets will be there Monday morning. That's our biggest problem, not that the assets are not there.

So any bank run being unfortunate is fine for us from an asset side. Of course, if the assets are not there, the likelihood of people trading, would go down, et cetera. So it's not that we would cheer for bank runs, but it doesn't cause any concerns for us except that the potential time gap. So in terms of proving cryptographically asset ownership, I know you guys initially released a year list of wallets and then the next step is, of course, you know, demonstrating that you control them. How are you going about doing that? So the simple wallets have been signed. We've said this is a Deribate address always opens something like that.

We have some catchphrase which is signed and people can, and we will publish this shortly. People can verify this. So there's Ether scan, for example, you can verify a message and you can see that we've signed our wallet with this message. So you use the privacy to sign a message on the particular address and then. people can verify with a certain signature that we will make available that we've actually signed this. So that's the simple stuff. The more complex stuff is fireblocks. So we can't easily sign with fireblocks.

But firebox is our whole wallet. So you can see all the deposit and withdrawals coming in from there. And we were thinking about sending a particular message at a particular time. So a transaction with a certain value. We were thinking about so the deribet, but then the numbers, the letters converted into numbers. So the D would be four and the E would be five and then there a bit spelled in a certain 4.5 BTC, whatever, something like that. We also have Cigdom, which is Swiss Bank. We keep assets there.

They don't have a way to sign. So signing is not possible for all our wallace, but for the wallace that we can't. sign we will show proof of control in another way, like a message at the particular time slot. And that way we will show control of all the assets. That's the idea. Yeah, that's clever. So what's the term for that steganography or something? Is it basically... It's in the end, no one can send that particular transaction from that wallet at that particular time but us. So we would, and it's possible, I think for being seen is to send to your own address

as well, which might seem strange, but you can send from an address to the same address. So we don't have to move assets. It's simply stays where they are, but it results in an on-chain transaction. In the end, that's you would be the only one looking at it and a few people like you, and they would click the links and would see that the messages are there. And of course, the bigger wallets are more important than the smaller stuff. But at least if you see that it's in control, and auditors would demand the same. So any auditor would want to see that we control those assets. I think the very first version of this was way back in 2011, Mount Gox did a transaction,

which was 424,000 Bitcoin, 424.44.24. So it was 4242424. Of course, the irony there was that was just a proof of assets, but they, I believe, it wasn't solvent for years. So that's what I meant in the beginning. So it's not only this. It's also the financial health. It's everything together.

So you want technical security. You want asset security. You want, we're getting a license. So you want regulatory security. It's like multiple pillars. And the only thing that matters, so no one cares or no one looks at the market at this moment, it's a bit without direction.

Voluntity is low, but people care about. trust. People want to make sure that instead of trading on 20 platforms, they will trade on five, and they want to make sure that they completely trust those five. So since, since FDX, we've seen a ton of judilious requests and forms to fill and all people want to discuss what way we do certain things, which is for them to make sure that they're as, they're comfortable with the risk they're taking. And it's more than proof reserves. It's, it's proof of many things. And that's the way forward. Yeah.

And, you know, people sometimes criticize proof of reserves. They say, well, it's not an assurance that there may be other liabilities, which are not incorporated into that. And, you know, you may not be privileged and liquidation, you know, depositors may be unsecured creditors, things like that. So I agree. It's, it is a patchwork. Yeah.

But the proof reserve is an excellent start. It's an amazing start. So I think it's way better than any traditional setup. has this far. And people, I think, overstates the audited financial because that's like delayed. This is real time. This is for us, it's daily. For others, it's weekly or monthly or whatever, but it's way more accurate and then outdated results, especially if you're, if you're publishing results for year end, let's say in May, it could be one and a half years later. And it's, Yeah, this is real time.

And I think that's what we're heading. That's the benefit of blockchain and of this new technology that we can do stuff like this. And that we didn't do this on day one. Of course, it's an evolution. And of course, they have some problems to make people realize the added value of this. So on the liability side, I think yourselves and Bitmex are the only platforms I've encountered so far that have done the full liability release. And then others have done either the Merkel sum and then some have just done very weak.

piecemeal releases where it's really unclear what really is being released. So tell us about how you were able to release the liability set without sort of leaking too much user data because of course, you know, historically the problem was when Coin Floor did it, the problem was you could look at specific accounts and see them over time, you know, because they might have had precise balances. How did you address that? The main way to address this is by cutting the balance in random pieces. So if you, I was just checking a random one that has like a thousand BTC, and we cut that in 50 pieces or 53 pieces.

If you have a small balance, so I just, another one I just checked, that has 0.2 sole, but it's also cut in two pieces. So even the small stuff is cut in smaller stuff, and the big stuff has got a multiple chunks. So in the proof of reserves, you would see a full list of liabilities, but the thousand would not show a thousand. The thousand would simply show, whatever, 10.4735 plus all these smaller numbers, all random numbers and random hashes.

So in your account, so if you look into Derbid, you see a list of all these 50 randomized, randomized asset counts and you see all the 50 hashes which and we also show the code how to identify those are unique and that they're yours and you can use it's an annoying job of course but you can look at all these 50 hashes and look at the overall liability file and see whether those assets are actually there so for the end user you can have the assurance that your aggregate balance is included in the many leaves and then from the third party perspective which I really like is a third party and not a client of the exchange, I can see all of the aggregate values there, but I can't at all associate those individual balances to specific users.

No, there's no way you can, you can say this, this is mine or this is whatever, certain firms assets. You can't see that. And there's no big values either. There's no wallet addresses. There's no nothing to identify a client. So there's no client details. There have been some discussions, in fact, dating back to 2015 about is there a knowledge approach to the problem, which is more black boxy, more technically complex? Vitalik had a recent blog post on this. Was this something that you guys considered at all? Was aside from, you know, you have privacy preserving approaches here, but is there a knowledge approach which you wouldn't actually have the full release of liability? Yeah, we looked at that.

We discussed it with our team and we decided this approach was better, or at least we thought this was smarter, because we liked the full transparency idea so that we say this is it. We can't be more transparent than we are. So this is like 100 or over 100% transparency. And secondly, we wanted to keep everything in-house and not so the zero knowledge solution would require a vendor. implementation and we purposely didn't want that. So we think this involves the core of the platform. This involves access to a lot of databases and key information, which we never share. We didn't want to share in this case either.

I think that's sensible. So some of the, it's interesting, you looked at Proved Reserve space, it's kind of like two main directions. One is like to go quarterly or biannually and have an audit firm come in and oversee the whole procedure. and then maybe because you have an audit firm, you know, while the assistance of the argument made, the exchange can maybe be more judicious about what they're sharing. And then the other approach, which is kind of used yourselves some bitmax is more, well, we're going to do it on high frequency with lots of transparency, and maybe because of that we don't need the oversight.

I know you mentioned that you're going for kind of a financial statement audit. have you contemplated trying to bring in an audit firm as well to oversee the proof reserve attestation? Yes, but we don't see the needs. So actually the entire idea of trust us is, or at least in our opinion, way more powerful in the absence of the need for someone to verify. It's not that we give some auditor a look under the hood and they can say it's fine, because you have to rely on the, let's say, the technical quality of the auditor.

the fact that the exchange might have not been open or have shown whatever subset of information, it requires trust in the process, it requires trust in the auditor. And I think that goes against what we're trying to do here. We're trying to put in trustless society. That's what that's what's everyone aimed for, right? That's the idea of Bitcoin, of blockchain. It's a technical solution that does not require a third party to approve, or whatever to review.

And that's why we chose this model. It's trustless. You don't have to trust us. You don't have to trust our front end. That's why you made the full code available. You don't have to trust an auditor. You don't have to trust anyone. It is the way it is.

This is full transparency. Now, I'm not asking you to be critical of other platforms, but what would you like to see in terms of improvements from your peer exchanges in the industry? I think in the end, frequency is important. So why do it monthly if you can do it daily? There's a reason for that. So if you publish it the first of the month,

you can deviate perhaps the rest of the month as long as it matches the next first day of the month. So I would be in favor of doing more frequently. I think it's important to do it for your biggest assets. I can imagine it might be too much work for small stuff, but it can be done for more than BTC. There's no real reason not to do it less than for top 10 coins instead of just a big name. So I think it should be covering more assets. And I think the full liability versus the Mercantree approach, I can imagine that it results in practical problems.

If the file size gets too big, you're not creating gigabyte files on a daily basis. that results in complexity for the user instead of actual benefits. No, you can't just easily open these files anymore. So I can imagine that some prefer the tree instead of the full list, but also the tree has its downside. You can hide negative balances. You can hide errors in that. Full liability is cleaner.

But I also understand that it's perhaps not for everyone. So I don't object to people doing that. But I do think the frequency and the number of assets is something people can improve. The common objection is window dressing, of course, which is basically borrowing funds on a short-term basis to satisfy a specific point in time at a station. With the daily frequency, I would argue that's not a consideration. It's hardly possible, but you would also see those assets move perhaps between the 24-hour slots on change. So you would see it.

Yeah, that's the other, that's my response is, well, the wallet addresses are known. So if there was window dressing occurring, it would be the easiest thing in the world for anybody to actually expose it. And people thought that was happening with Gate I.O. Actually, in my opinion, wasn't the case, but there was a large transfer. Not exactly during the time when they did the proof reserve attestation, but yeah, there was a large transfer that people noticed. So that's my point is these are very transparent. You know, we can see what's happening. So if there was anything sketchy at the point where they've released the wallet addresses, it would kind of be noticed.

Yeah. I don't worry about that too much. This is, as I said, I did my review. And from my first glance, you guys performed extremely well. And you did two things differently, maybe even more from everybody else, which is the daily attestations, which I think is really praiseworthy. the full asset coverage, and then also having the end point regarding margin, open interests, especially for an exchange like yours where that's very salient. So top scores from me, thank you for doing it.

Thank you. Anything else you want to mention before we hop off? No, I appreciate all your efforts. So we mentioned already it might be boring to review all of them. And the devil is in the detail, but you've done a great job and appreciate the recognition. Well, I mean, it's recognition of hard work done. So keep it up. I'll certainly be validating these on an ongoing basis.

And yeah, Luke, thank you very much for joining us today. Thank you.